Stars with Plumbing Issues: The Multi-component Ejecta of Common Envelopes

The gist The formation of a circumbinary disk with collimated outflows surrounding the central binary arises naturally from hydrodynamical simulations of CEE, and that their morphology and kinematics are consistent with observations of WFs.

Common-Envelope Evolution Context

Common-envelope evolution (CEE) is a brief phase in binary stellar evolution where a giant star engulfs its companion within its envelope, causing orbital energy and angular momentum to be extracted, which is responsible for forming close binaries. CEE proceeds through three main phases: the loss of orbital stability, a dynamical plunge-in, and a slower spiral-in. Because of these distinct timescales and physical processes, self-consistent 3D hydrodynamical simulations cannot currently model CEE from start to finish and must instead focus on one phase at a time. The inherent lack of direct observations of ongoing CEE further complicates its modeling, necessitating reliance on post-CEE observations like planetary nebulae (PNe) or Water Fountains (WFs).

Water Fountains as Observational Constraints

Water Fountains (WFs) are an ideal class for comparison because they likely capture a snapshot of the envelope ejection while it is still happening, and they are identified by their collimated water masers. WFs are younger objects with dynamical ages of at most ∼ 200 years, and they host an equatorial torus containing a substantial fraction of the original envelope mass. The authors argue that CEE is the only known event capable of explaining all observations above, as WFs provide constraints on envelope ejection during the spiral-in phase.

Simulation Methodology and Numerical Sensitivity

The simulations analyzed a subset of 3D CEE simulations from Borges, using the moving-mesh hydrodynamics code MANGA. The choice of numerical solver significantly impacts results, as seen when comparing the HLLC case to the HLL case, where the HLLC develops a pronounced bipolar structure in edge-on views while the HLL does not. Furthermore, simulations show that the outcome of CEE can be highly sensitive to numerics and to the final separation reached after the plunge-in phase. The HLLC run maintains an approximately axisymmetric morphology throughout evolution, whereas the HLL becomes asymmetric, and its treatment of contact discontinuities allows collimated outflows to form and remain coherent for longer periods.

Formation of a Circumbinary Disk (CBD)

The paper shows 7 simulations of CEE with an RGB during the plunge-in phase and onset of spiral-in, identifying the formation of a CBD. The transition between the plunge and spiral-in phases is defined by the condition aP/a˙ < ϵ, but morphological indicators suggest that if a CBD forms, the transition is qualitative rather than just operational. The thick CBD formed shortly after the plunge differs from post-CEE thin CBDs because it remains transient and may evolve into a thin disk on timescales of a few centuries. The radial scaling of density shows that beyond a certain radius, the envelope is no longer well represented by a Gaussian profile, indicating two distinct components in the ejecta: a thick disk and an expanding component.

Collimated Outflows and Comparison to Observations

The simulations show that collimated outflows can arise from purely hydrodynamical processes without requiring MHD effects, consistent with previous findings. The opening angle of the outflow increases slightly over time, evolving from ∼ 30◦ to approximately 40–50◦, which is closer to those inferred for WFs. The maximum speed of the ejecta begins at approximately 250 km s−1 and gradually decreases to about 200 km s−1 during evolution, velocities comparable to the escape speed of the central binary.

Insights for Future Work

The paper suggests that WF sources exhibit outflows velocities that increase with time, which may reflect a decreasing binary separation. The characteristic outflow velocity in WFs increases as the system evolves, potentially due to a decrease in orbital separation or envelope ejection powered by orbital energy release. Furthermore, the differences between WFs and other post-AGB systems with thin disks may reflect the evolutionary stage at which CEE occurred. The most robust observational signature of a CEE event may be the wide collimated outflows, formed due to the redistribution of angular momentum in the still-bound envelope and formation of a CBD.

Conclusion

Both numerical simulations and observations point toward the same scenario for the spiral-in phase: the still-bound envelope redistributes into a thick CBD, and is then preferentially ejected in the polar direction through collimated outflows. The results presented may provide useful initial conditions for future analytical or numerical modeling studies of the spiral-in phase of CEE. The collimation in WFs likely arises from geometric constraints imposed by the thick equatorial disk, forcing the ejected envelope into collimated polar flows.

1. Bold header: Analytical Initial Conditions for CBD Formation

The improved AI system can generate an analytical description of the properties of the CBD that can be used as initial conditions for future simulations, providing a structured starting point based on Section V, which notes that our simulations indicate the formation of a CBD rather than a spherically symmetric expanding envelope.

2. Bold header: Outflow Morphology and Kinematics Matching

The system can compare simulation results with observations by assessing how well the simulated collimated outflows match Water Fountain (WF) morphologies, specifically noting that the opening angle increases slightly over time, evolving from ∼ 30◦ to approximately 40–50◦.

3. Bold header: Physics-Informed Numerical Sensitivity Assessment

The improved system can quantify the impact of numerical solvers by identifying which components are most sensitive; for instance, it can determine that numerics are the highest source of impact in the formation of such structures, highlighting that the HLLC solver is crucial for capturing bipolar features while noting that the more diffusive HLL solver tends to smear out such discontinuities.

4. Bold header: Phase Transition Modeling Refinement

The AI can refine phase transition criteria by suggesting a stricter threshold, arguing that it is better for the threshold to extend slightly into the spiral-in phase rather than into the plunge, based on the observation that the physical transition happens at 200 days in specific simulations.

5. Bold header: Outflow Velocity and Orbital Decay Correlation

The system can model a feedback loop where an increase in velocity from tens to a few hundred km s−1 corresponds to a contraction from separations of order tens of R⊙ to a few R⊙, linking the observed acceleration of WF outflows directly to the gradual orbital decay caused by the CBD.

6. Bold header: Evolutionary Stage Discrimination

The system can develop classifiers that distinguish between different evolutionary stages by correlating morphology with stellar evolution history, proposing that the differences between WFs and other post-AGB systems with thin disks (such as the Red Rectangle) may reflect the evolutionary stage at which CEE occurred.